C.F.I. OCCASIONAL PAPERS

A SECOND LOOK AT AGATHIS

by

MoR o Bowen and ToCo Whibmore1980

A SECOND LOOK AT AGATHIS

M.R. Bowen and T.C. Whitmore

Contents

INTRODUCTION

TAXONOMY

FORESTS WITH AGATHIS AND THEIR EXPLOITATION

PlANTED AGATHIS

PROPAGATION

PhenologySeeds and their storageVegetative propagation

PESTS AND DISEASES

PestsSeed eating moth AgathiphagaOther insect pests

Diseases

PRODUCTS

TimberThe resin industry

SEEDLING RESPONSE TO LIGHT

INTERNATIONAL SPECIES/PROVENANCE TRIAL

REFERENCES

ACKNOWLEDGEMENTS

1

1

1

2

3

8

11

12

15

17

18

INTRODUCTION

Since the publication of the 'First Look ~ Agathis' (Whitmore, 1977)more information has come to light on many of the topics covered there;several papers have been published and seed has been collected a~d dis­tributed as an international species/provenance tiial.

The Oxford-based investigation of Agathis is summarized in Bowen andWhitmore (1980), \Vhi1e in this paper we publish the new details notappropriated to that summary, and using the same .category headings asin Whitmore (1977). These data in conjunction with those publishedpreviously provide an up-to-date statement of knowledge of the genus.We then go on to give details of the provenance trial and a note on apre1~inary experiment on the response of seedlings of different speciesto three shade conditions.

TAXONOMY

Since the First Look was written a monographic reV1S10n of the genus hasbeen undertaken and is published as Whitmore (1980). ~ere are thirteenspecies of which two have two subspecies and one of these is the onlynew taxon discovered .. An account of Agathis for Flora Malesiana, coveringSumatra to the Bismarck archipelago, is in' the course of preparation byDr. D. J. de Laubenfels. He is likely to take a narrower concept ofspecies delLmitation and make many or all of the montane populations of~. dammara in Borneo into different new species (including two of over-lapping elevation on Mt. Kinabalu) and to split in two the -lowlandspecies of Celebes, Philippines and Moluccas. Foresters will have tojudge for themselves whether this 'splitter's' view of the genus isjustified.

FORESTS WITH AGATHIS AND THEIR EXPLOITATION

Additional notes on wild Agathis and its exploitation have come to hand.The overall picture on its ecology summarized in Whitmore and Page (1980)is not altered,. especially the generalisations about population structure.Many Agathis forests are difficult to reach, or soon to be exploited, anddata on them are recorded here.

Sabah. The montane stand of ~. dammara near Keningau, Mt. Kinaba1u hasnow been destroyed. It had poles but seedlings wer~ rare (D. Chai, pers.comm.). Dense, accessible stands (Which are probably of ~. borneensis)have been discovered in the Crocker Range at over 600 m elevation onsed~entary rock. In the lowlands a further stand on u1trabasics hasbeen found near Ka1abakan.

Sarawak. Miscellaneous notes about A. borneensis in heath forest are tobe found scattered through Brunig (1974) •.

Philippines (~. dammara). Additional general notes on range and ecologyare given in Brown (1951) and by S.C. Halos and B. Tuma1iuan (pers. ~.).Agathis occurs in a strongly seasonal cl~ate not just on Palawan but

- 2 -

also in Abra province, northern Luzon, where it occupies upper slopesfrom 800 to 900 m interdigitating with Pinus kesiya forest on thelower slopes. In the intermediate zone Agathis is reported to extendupto 2000 m elevation on Mt. Balarit (2300 m), but is absent from thetop 200 m of Mt. Balutitic (1800 m). The absence from ridge crests isgeneral in northern Luzon in contrast to Palawan. The largest treesare found towards the upper slopes and on south facing slopes which aremore densely populated than north facing slopes on a hill system thatruns with spurs pointing eastwards. An area of 196 000 ha is estimatedto contain 1.3 million m3 of Agathis with up to 13 000 trees over 0.6 mdbh in a 1000 ha area.

Queensland. It seems probable that A. atropurpurea and A. microstachyawere never cammon or widespread. Both are now very rare g D.I.Nicholson only knows of ten trees remaining of ~. microstachya andA. Gardner suggests that probably 100 remain in total.

The original geographical range of ~. atropurpurea and ~o microstachyais not exactly known. The ranges of these two and ~. robusta overlapand they do occur, but rarely, at the same place; for example~. atropurpurea and ~. microstachya are both at Lake Barine on graniteat 750 m, with A. robusta under 2 km away. Two large (c. 2 m) stems of~. microstachya-have been 14C dated to 1050 and 450 yea~s old.

In south Queensland A. robusta never occurs over 200 m elevation andmost is at lower lev;ls (pace Whitmor~ 1977). It occurs on podzols aswell as on more mesic soils.

Santa Cruz (Solomon) Solomon Islands (~. macrophylla). Exploitation ofthe Ndendo stands was completed in early 1979. Wildings and nursery­raised seedlings were line-planted with a high degree of survival. Thespecies is not endangered, but seed can now only be obtained fromscattered individual trees and small pockets which escaped logging untilthe new forest matures.

New Hebrides (~. macrophylla). The Aneityum forests were logged againfrom May 1978 to June 1979. Some 5000 m3 of Agathis timber were expor­ted during 1978 at c. £96 m-3 f.o.b. The species occurs as patches upto 1 ha in extent and 200-300 m distant from each other. It is concen­trated on slopes- protected from the prevailing southeast winds.

New Caledonia. At Riviere Bleue some 2000-3000 m3 are being cut perannum, mainly of !. lanceolata, for local use.

New Zealand. The Kauri Management Unit, Auckland Conservancy, publishedin 1978 a compendium of all measurements in second growth !. australisforests including volume increments. This very substantial body of datais ~portant for future silviculture of the New Zealand Kauri o

PLANTED AGATHIS

Java. A study of mineral cycling in the central Javan plantations wasconducted for 15 months to February 1978 by Drs o L.A. Bruynzeel.Publication is anticipated in 1980.

- 3 -

Costa Rica. Agathis robusta has done well in a trial conducted byTurrialba (Anon. 1976) on a site at 1220 m elevation on a leachedrecent volcanic soil in a rainfall of 3725 mm/year. At age 6.3 yearsthe following measurements were obtained; h 7.55 m (m.a.i. 1.27 m),d 11.58 cm (1.96), vol. 15.66 m3/ha and h 11.67 m (1.99), d 16.71 cm(2.88), vol. 88.76 m3/ha. These should be compared with the resultsshown in Figs. 5.1 and 5.2 in Whitmore (1977).

PROPAGATION

Phenology

The months of the year at which ripe female cones are present variesfrom country to country. In many cases the available information islimited to the period at which 'mature' cones have been seen and littleis known about the t~e scale of development, the peak collections period,or variations from year to year. Table 1 shows the revised data based onpersonal observations and the more reliable field reports, and replacesthe similar table in Whibnore (1977).

The development of the female cone of Agathis robusta south Queenslandprovenance was described in detail by Nikles (1965), see Whibmore (1977).It takes 18 months from emergence to maturation, and it is now thoughtthat other species differ.

In Fiji trees were examined in early January 1979 near Colo-i-Suva, inthe Nausori Highlands and at Nadarivatu. A few, very young, female coneswere seen, but none of intenmediate size. The bulk of the seed crop wasestimated to be mature by early to mid-February, with later conesmaturing in March. It would thus appear that a 12-15 month growth periodis required for this provenance of ~. macrophyl1a.

At Gumi in Papua New Guinea material collected in mid-February 1979 fromfue crowns of mature trees of ~. robusta subsp. nesophila showed largeexpanding buds which experienced tree cl~bers identified by eye asyoung female cones; dissection under a microscope confirmed thisopinion. These cones are ~aid to be fully mature by mid-November, sug­gesting a 9-11 month developmental period. Certainly no cones of inter­mediate age were seen on the large number of trees examined, lendingsupport to the idea of a short development period.

In contrast to this relatively fast maturation period N. Q. Zabala(pers. ~.), has tagged developing cones on two trees of !. dammaraof unknown (Philippines) origin planted at Los Banos, Luzon, and hisresults suggest that full development takes about two years.

On the Palawan Island, also in the Philippines, !. dammara occurs aspatches separated by up to 2 km from each other. It is reported toproduce mature seeds from mid-February to mid-April at the northern end,and from June to September in the south. The clUnate is less seasonalat the south tip of Palawan (Fig. 3.1 in Whitmore, 1977) which mayexplain this anomaly.

- 4 -

Table 1. The months when female cones mature (+ some, * most mature cones)

0 N D J F M A M J J A S

Fiji, Viti Levu~. macrophylla + * * +

New Hebrides, Aneityum~. macrophylla * * +

Solomon Islands-, Santa Cruz~. macrophylla * * +

Australia (!. atropurpurea,~. microstachya and!. robusta) all 3 spp. * *

New Caledonia!. corbassonii * *!. .lanceolata - + * +~. moorei * *!. 2Y!!!!. * +

Papua New GuineaSirinumu

!. robust:a - nesophila - + *Gumi

!. robusta - nesophi1a + *Sepik~. labil1ardieri * +

Philippines ~. dammara)Zamba1es, Samar *Zamboanga; Cebu ?Luzon, Los Banos *Palawan N - + * *Palawan S * * + +Luzon, Abra + * + +

Borneo, Balikpapan!. borneensis + * +

Java (!. dammara)Pasir Hantap + * +Situgunung + * + + + * +

- 5 -

At Pasir Hantap in west Java small plantations of three provenances ofA. dammara (two Celebes, one Moluccas) and of A. borneensis (Brunei);ere examined in mid-March 1979. Most trees b~re female cones in allstages of development from very small to almost mature, suggesting adevelopment period of 14-18 months and possibly a fairly extendedfruiting season. Local foresters say that the peak of seeding is Apriland numbers of viable seed obtained per cone vary noticeably betweenprovenances but is fairly constant from year to year. Viable seeds percone are 60-70 for ~. borneensis and the Moluccan provenance of~. dammara, and 15-20 and 5-10 for the two Celebes' provenances.

In other countries from Which data are available, New Hebrides and SantaCruz Islands (!. macrophylla) and New Caledonia (all five species), adevelopment period of 15-18 months is indicated.

In a small plantation of !. moorei in New Caledonia, H. ,Gay (pers.reports that seed-set and the number of cones developing is betterthe leeward side of the area, and that while trees on the windwardusually bear a moderate cone crop, almost all the seed is sterile.variation is shown by the seed collected in February 1979 when thecone crop was harvested from seven trees.

.£.2!!!!!.. )onsideThis

total

Table 2. Total cone and seed collections from seven p1antation- growntrees of A. moorei

Tree No. of cones Weight of seed Average weight ofNo. collected obtained (g) seed per cone (g)

1 37 1935 52.3 leeward2 or 5 15 1048 69.84 18 907 50.46 22 356 16.1

10 12 313 27.120 15 3 0.225 7 35 5.0 windward

Considerably less information is available on the period of developmentand time of pollen shed in the male cone. Many of the cones continue toexpand on the tree after the pollen has been released and fall over anextended period.

Grafting

In New Caledonia, a seed orchard (12 years from grafting) of some 28 trees,has produced a moderate number of mature cones containing fertile seeds.Scions 10-15 em ,in length and 15 rom in diameter taken from mature foresttrees were side-veneer or top-cleft grafted on to 1 m tall juvenile rootstocks of ~. moorei, the top-cleft grafts producing a firmer union.~. ~ and !. 1anceolata grafted onto !. moorei rootstock have grownwell but the biggest grafts are thought to be A. lanceolata on root stocksof the same species. -

At Atherton, north Queensland, mature scions of A. robusta have beensuccessfully grafted onto juvenile rootstocks of-both ~. robusta and

- 6 -

~. microstachya. The reciprocal grafts are also healthy.

Seed storage

Viability studies initiated in 1976 by R.D. Smith at the Royal BotanicGardens, Kew, continue to confirm that Agathis australis, !. macrophyl1aSolomon provenances and ~. robusta seed, when conditioned to moisturecontents between 16 and 6 percent (wet weight basis) behave in an"orthodox" manner, .!..~. reducing the moisture content of the seed, orlowering the temperature at which it is stored, prolongs longevity. Itappears that ~. macrophy11a seed possesses a lower inherent longevitythan that of~. robusta which in turn is lower than that of ~. australis.One unexpected and unusual feature is found in the behaviour of~. macrophy11a which wholly or partly accountg for its low viability.When seeds are stored at temperatures below 0 C at moisture contents below16 percent a proportion is killed ~ediately. The rest lose their via­bility in the manner which would be predicted from the behaviour of seedat temperatures above zero Q The proportion of the seed which is killedon exposure to sub-zero temperatures increases with a decrease in eitherstorage temperature or moisture content.

Measurements made on ~. macrophy11a Santa Cruz provenance showed that themoisture content of seeds from freshly shattered cones is £. 65 percent ona wet weight basis. Provided that the moisture content does not fall toor below 20 percent through natural drying, and the seeds are supported ina thin layer on wire mesh and sheltered from direct sun and rain, theyremain completely viable for at least one week under ambient conditions.The rate of drying under natural conditions is roughly inversely propor­tional to daily rainfall. At moisture contents at or below 20 percentmeasurable losses of viability occur. These field measurements agreewith extrapolations from earlier laboratory experiments at lower temper­ature and moisture contents o No immediate loss of seed viability wasfound if unripe cones were left in direct sunlight for up to four daysto induce shatteringo

Seeds which had been held at over 20 percent moisture content underambient conditions for four days before being despatched by airmailarrived in Britain fourteen days later with 90 percent viability. Whenthese seeds were then dried at 160C and 14 percent relative humidity forfive days the moisture content decreased to 6 percent and viabilitydropped to 75 percent.

From these observations it was concluded that it should be possiblesuccessfully to transport fully viable seed to Britain (or elsewhere)where facilities are available to rapidly dry them and so prolong theirlongevity so long as moisture content is over 20 percent. On the basisof these conclusions the decision was made to collect seed for theinternational species and provenance trial described later. Unfortu­nately much lower viability than predicted was obtained in the seedcollected for the trial.

Despite the inference which could be drawn from the previous viabilitystudies, no correlation was found on arrival in Britain between the

- 7 -

germinability of the seed collected for the trial and its moisture con­tent. It must therefore be deduced that at least one further factormust be controlled if Agathis seed is to be safely collected and trans­ported. Two possible factors can be identified, the carbon dioxide/oxygen balance and chilling injury in transit.

Of the seed collected for the species and provenance trial the mostviable was a small sample kept and transported in a substantiallymoistureproof, and therefore gas-tight, plastic container. The twobatches of ~. macrophylla seed previously successfully transported toBriain were both at lower moisture content and both very densely packed.This suggests that the e02/o2 balance may be involved. It has not beenpossible to carry out the necessary exper~ents to confirm or deny thispossibility.

Secondly, the seed might suffer 'chilling damage' in transit. The normalmin~um temperature in the cargo hold of a jet airliner is 7°e butexceptionally it may fall to zero. Both early batches of ~. macrophyllaseed which reached Britain with high viability were in fact close-packedin insulated containers. The seed of various species which died weremore loosely packed and poorly insulated. As this seed was also wetterit may have succumbed to 'chilling injury' at high moisture content.The solution to this problem would seem to be to transport seed in wellinsulated containers. Further experiments are planned.

Many other foresters report difficulties in storing Agathis seed for morethan a few weeks (~.a. Suriamihardja, 1979) and experimental results arerarely published. However, between 1958 and 1962 Queensland State ForestService stored air dried seeds (12-15 percent moisture content) of'~. robusta' (possibly in fact ~. microstachya or ~. atropurpurea) at-lSoF in airtight containers with no loss of the viability of c. 64 per­cent. In non-airtight containers viability fell to 17.5 percent, whileat room temperature all seeds died. These results are in broad agreementwith those of R.D. Smith.

Early results from New Zealand on the storage of seed of ~. australis,the least 'tropical' of the genus, indicated that the seeds of thisspecies also posed long term storage problems (Hocking, 1935; McKinnon,1936; Sand, 1936). Hutchin's (1918) assertion that ~. australis seedwill remain dormant in the ground for many years before germinatingremains unproven but would not in any case appear to hold true for theAgathis species of tropical rain forest, where numerous pests are likelysoon to destroy non-germinating seeds. By 1946 MacMorran (1946) hadshown that seed of ~. australis stored at ~. -lOoe remained 64 percentviable after two years, while seed held at room temperature died. Andin 1957 Mirams demonstrated that seed moisture content during storageis also of great importance. Recently Preest (1980), in a comprehen­sive experiment lasting almost twelve years, studied the combinedeffects of both temperature and seed moisture content on longevity.Preest's (1980) results are based on only one exper~ent and one seedlot but they agree with and extend those of R.D. Smith and fromQueensland, and are by far the most comprehensive data available. Seedbatches of A. australis were stored in air-tight containers at moisture

- ° °contents of 6, 10, 15 and 20 percent, and at temperatures of -10 , 5 ,10°, 150 and 200 e. Longevity was dependent on both factors althoughmoisture content was the more critical. Preest concluded that 'the

- 8 -

optimum moisture content appeared to be around 6 percent. For short­and medium-term storage (upto 6 years) the opt~um temperature was about50 C. For larger storage, or higher moisture content (10 percent), theresults suggested that the opt~um temperature is below freezing point'with viability little ~paired, and probably in excess of 12 years.

All the available data thus indicate that satisfactory seed storagetechniques are now being worked out for some species. However, furtherwork is needed to extend these results to' tropical species and to developbetter means of transp~rting seeds from country to country.

Vegetative propagation

Work on the vegetative propagation of Agathis has continued, at theInstitute of Terrestrial Ecology, Ediriburgh (Leakey and Longman, 1978),to examine the variation in rooting due to cutting origin. Cuttingswere taken from the main stem and lateral shoots of seedlings, and frompreviously rooted cuttings of ~. macrophylla. All cuttings were treatedwith a 'quick dip' in 0.22 NAA/IBA in methanol, and set under mist.

Cuttings of main stem origin rooted better than those from lateral shoots,the most pronounced differences being in the rate of rooting during thefirst five weeks. Subsequent growth of these rooted cuttings confirmedobservations that lateral shoot cuttings tend to grow plagiotropicallywhereas main stem cuttings are orthotropic.

Cuttings of the indigenous Agathis of Peninsular Malaysia (!. borneensis)have been successfully rooted at Kepong (Momose,1978). Cuttings 150-200mm long including severa·l nodes and with leaves left attached on the upperpart were successful from coppice shoots but not from the branches ofmature trees. Short cuttings of one leaf and bud, similar to thoseused at Edinburgh, successfully rooted after six months (these two'were presumably from coppice shoots). It should be noted that Agathishas not previously been known to coppice.

A mist propagation unit has recently been built at Oxford and furtherinvestigations on the rooting of cuttings, using a number of Agathisspecies, have heen started.

Rooted cuttings~. 0.2 m tall of !. dammara Java provenance,!. macrophylla Fiji provenance and ~. robusta raised at Edinburgh asdescribed in Whibnore (1977) were successfully introduced to PeninsularMalaysia by air in October 1977 and, after a period in a nursery, plantedout in the open at the timber concession of Perkayuan Tenggara at BukitIham. Initially!. macrophylla Fiji prove~ance grew best. In mid 1979all were reported to he growing vigorously (P.F. Burgess pers. ~.)though same still remained plagiotropic.

- 9 -

PESTS AND DISEASES

Pests

Seed eating moth, Agathiphaga (Lepidoptera: Aglossata; Agathiphagoidea)

New information has now been gathered on this serious pest.

Fiji (~. macrophylla). In general, local foresters report that the levelof infestation by Agathiphaga vitiensis is low, from 0.15 percent andvarying between areas. However, at Nadarivatu, several cones, estimatedto be a month from maturity, were split open and all contained larvae.The infestation involved from 20-30-(70) percent of the seed. Aboutone third of the infested seed had two holes drilled in the corners,while one tenth had three hole's - the third being centrally placed(andperhaps not Agathiphaga). A few seeds had four holes but theremaining infested seeds differed little in external appearance fromunattacked seeds. A very small number of the seeds were found with oneor two holes in the wing but with no apparent damage to the enclosingcone scales. None of the cones opened at:Nausori Highlands containedattacked seed.

New Hebrides (~. macrophy1la). Cones collected on Aneityum island weremature and contained 110-180 seeds, but up to 95 percent of these wereempty. In all cases two or three of the remaining seeds were infested.The numbers were proportionally higher in cones with more developingseed. Usually four holes were found, but more rarely two, at the cornersof attacked seed. Where only two holes were present they were invari­ably adjacent on one of the short sides of the seed, and never situateddiagonally. In fact no diagonally placed holes were found throughout thedistribution range. The holes were sometimes drilled only through onecoat of the testa, and sometimes completely through the seed.

A small number of cones were found with 1.5 rom diameter holes in thescales; in these instances the central 'candle' . often showed frassdamage and the cone was distorted. A Nitidulid beetle was found inone of these cones.

Solomon Islands (~. macrophylla). High levels of Agathiphaga damage arereported each season in collections made on Ndendo '(K.D. Marten andT. Nolan, pers. ~.) and these severely limit the quantity of seedavailable for replanting.

New Caledonia (A. moorei, A. corbassonii). Several hundred cones ofAo moorei were ;xamined, c~ntaining many thousand mature seeds; ofthese only 20 seeds were found to be destroyed by the larva ofAgathiphaga. All had four holes per seed. Sfmilarly, only three seedsfrom 85 cones of ~. corbassonii were attacked. This appears to be the'first record of Agathiphaga in New Caledonia, although H. Gay (perso~.) has noticed a small number of damaged seeds in previous years.No seed damage was seen in the small number of ~. lanceolata conesexamined, although there is no reason to suppose that this and the twoother sympatric species (~. montana and ~. ~) are LDmune to attack.

- 10 -

Papua New Guinea. Although pests appear to attack the cones and seedof Agathis in Papua New Guinea (see below) there is, as yet, no evidencefor the presence of Agathiphaga.

Australia. The original description of Agathiphaga queens1andensis wasmade by Dumbleton (1952). Since then, however, no one from theQueensland State Forestry Department has reported infested seed butB.P.M. Hyland (pers. ~.) reports that larvae can be found in mostyears in seeds of ~. robusta growing in north Queensland. There arenever any holes in the seeds. It seems reasonable to suggest that seedsof~. microstachyaand!. atropurpurea could also be attacked o

Other insect pests

Borneo o One recently fallen female cone was found to be heavily attackedby pin hole borers. Damage was mostly confined to the seeds and central'candle' and there were few holes through the scales. GoS. Robinsonidentified the larvae as a member of the Ge1echiidae.

Java. A study has been made of the termites associated with A. dammaraplantations in Java by Z.B. Kiman (BIOTROP Annual Report 197677)0'About 15 termite species of 3 families were recorded. Two speciesbelong to the family Rhinotermitidae, i.~. Schedorhinotermeslongirostris (Bauer), and~. javanicus (Holmgren). More than 12 speciesbelong to the family Termitidae, namely: Microtermes insperatus Kemner,Macrotermes gilvus (Hagen), Nasutitermes sp., ~. acutus (Holmgren),Odontotennes spo, O. sundaicus Kemner, O. malaccensis Holmgren,Q. sarawakensis (Holmgren), Q. javanicu; Holmgren, Capritermesbuitenzorgi Holmgren, £. spo, and £. ceylonicus Holmgren. Only onespecies belongs to the family Ko10termitidae, l.~. Ka10termes pintoiKemner. Schedorhinotermes longirostris and~. javanicus were the mostprevalent species within plantations in Baturraden, Situ Gunung andGunung Walat. The percentage of occurrence of ~. longirostris was81.25, 48.14, and 54.55 percent in Baturraden, Situ Gunung and GunungWalat respectively; whereas that of ~. javanicus was 12.50, 25.92,and 13.64 percent respectively in the same areas. Severity of damage,age of trees attacked and the altitude of each plantation were alsorecorded.'

Papua New Guinea. Two species of moth have been raised by H. Roberts(pers. ~.) from cones of Agathis robusta subsp. nesophila andA. labillardieri collected in Papua New Guinea. The most common has a;ingspan of 18 rom, with white fore- and grey hind-wings. At rest thewings are held at an acute angle; the body is yellowish with pairedblack spotso This has been identified by J.D. Bradley as possiblyTarphyscelis sp. (Yponeumetidae) and is widespread from Port Moresbyto the east Sepik and New Britain. The second moth, found only atGumi near Bulolo, is smaller, with a 12 rom wingspan, of the samecolouration with the wings held flatter o It has been identified byJ.D. Bradley as an un-named genus near Prose1ena (Tortricidae).

A damaged cone collected by J. Zieck at Sirinumu, Papua, was examinedby H. Roberts. Its external ~ppearance coincided with green conespreviously seen from other parts of the range. Lepidopterous larvaehad moved through the scales as well as in the central 'candle'. Manylarvae appear to pupate just inside the outer surface of a scale prior

- 11 -

to emergence through a 1.5 rom diameter hole, in or between, scales. Incases previously examined by H. Roberts a number of larva have pupatedin a white silk cocoon on the cone surface or stalk. Moths emergearound mid-February from green cones collected in mid-November, l.~.

at a time when the mature cones would be expected to have fallen.

Bud material collected from the Gumi area was also examined. Thelarger buds, identified by climbers as potential female cones, allshowed signs of damage by lepidopterous larvae and two contained livingspecimens. All the buds were more or less completely destroyed so anattack at this stage is potentially serious. Frass and damage was alsonoted on fresh young stems and in slightly older, though still green,wood, suggesting the larvae could move freely to surrounding buds. Inyoung cones pupation probably takes place outside the cone on nearbystems. H. Roberts suggests that these larvae are very probably of thesame species as those attacking the adult cone, io~o ?Tarphysceliso

The frass damage seen in the central 'candle' and the holes boredthrough the cone scales look somewhat alike in cones collected in theNew Hebrides, Atherton, Australia, at Balikpapan, Borneo and in PapuaNew Guinea although it is probable that the larvae causing the damageare different in each place.

No large borers in standing timber have been reported in Papua NewGuinea (H. Roberts, pers. ~.) although Ambrosia beetles and barkbeetles peculiar to Agathis rapidly invade felled logs. Roberts con­siders that Agathis is likely to have many species not previouslydescribed.

Australia. D.I. Bevege (pers o ~.) reports the presence of the larvaof a small moth (adult wingspan 9 mm) in seeds of A. robusta. So farthe species is unidentified, but may be related to-the Hoop Seed MothHieromantis (Stathmopodidae) of Araucaria cunninghamii. Damage isdifficult to see except as tiny punctures in the seed wing, and aspin-head holes in the seed scale, made as the larva works its waythrough 10-12 seeds.

A new undescribed species of weevil, similar to the weevils of Araucariafound throughout its range, has also been found in the cones of A. robusta.It is possibly a member of the Nemonychidae but has no known effects(D.I. Bevege, pers. ~.).

In Atherton public park one ~. robusta tree with many mature cones wasexamined. One mature cone, still on the tree, had a 1.5 mm shot hole,the outer edges of three scales were eaten and an active 10 rom whitelarva found (see Papua New Guinea).

Diseases

Halos and Quinones are exam1n1ng the correlation between diameter classand the presence of wood decay in Philippines Agathis. The rot isthought to be caused by a species of Fomes and fruiting bodies have beenobserved on apparently healthy as we1r-;;-dying and dead trees. SUni1arfruiting bodies are also found on neighbouring~ kesiya. Theseworkers say that there is a strong correlation between diameter and wood

- 12 -

rot, especially in trees over 1.6 m diameter. They have classified thedecay into three types:-

Type AB

c

rot extending from the bark into the cambium.rot starting from the root and extending upwards intothe stem to form a hollow heart.rot also originating in the root but decayed regionsappearing only as scattered spots across a section.

The type of rot occurring varies from region to region .and may be cor­related with climate. Type A is found in northern Luzon and the Apayao­Ka1inga region of central Luzon, where rainfall is compressed into asix month period. Type B (with some type C) is found in northern Samarwith a heavy, evenly distributed rainfall; while Type C rot has beenfound at Zamboanga (Mindanao), with an evenly distributed though lightrainfall. The majority of trees examined in the Abra area of northernLuzon were between 1.2-1.7 m dbh, although very occasional trees wereestilnated to be 2.2 m dbh. Quite large numbers of trees above 1 0 8 mdbh were dead, although still standing 0 Large dead trees were alsonoted on Palawan and resin tappers report that frequently tapped treesdevelop a rot in the older cuts. Termites also enter tapping wounds andultimately lead to death.

PRODUCTS

T~ber

In general, New Zealand Agathis has a better reputation in the trade fordurability. T~ber of ~o macrophylla Vanikoro provenance is said to havea higher resistance to soft rots than most other species (CSIRO ForestForest Products Newsletter 374, July 1976). The more 'tropical' speciesof Agathis, especially Indonesian, show a tendency to possess growthrings whereas Australian species have none (W. Smith, pers. ~.).There are no significant timber differences between the three Australianspecies except probably in the amount and extent of spiral graing Sprialgrain is also more prevalent at high altitudes.

Some provenances or species may have a heritable tendency to spiralgrain. This is of special importance in such a low strength wood, and

.is clearly an important character to avoid in selecting trees for aprovenance trial or seed orchard. As far as possible any trees with aspira1ity greater than 5_60

, in a 1.2 m length should be rejected(W. Smith, pers. ~.). Smith believes ~. macrophylla from the NewHebrides and ~. ~ have higher degrees of twisting than other species,with ~. ~ the worst.

Within the Philippines, trees from Abra (N. Luzon) are of very twistedform with the spiral anticlockwise and branching is much 'heavier' thannormal. Elsewhere in the Philippines form is excellent (S.C. Halos andA. Quinones, pers. ~.).

The resin industry

In the Sepik region of Papua New Guinea, a good tree will produce 20 kgof gum per year, at a world market price for top grade resin of £300 pertonne.

Percentage increase in height growth of three species and five provenances of Agathis grownin a glasshouse under three light regimes. a. unshaded conditions; b- f differing light conditions.

0--0 unshaded ..... Iight shade ....... heavy shade

160 160a. unshaded conditions b. ~ macrophylla Santa Cruz

120 120

80 80

40 40

0 00 40 80 120 0 40 80 120

160 160c. ~.macrophylla Aneityum d. ~. macrophylla Vitu Levu

Q)C/)

co 120 120Q)'-0c

:E 80 80C)

Q)s::.Q)Q) 40nsCQ)0'-Q)

0c- O40 80 120 0 40 80 120

160 160e. ~:moorei N.Caledonia f. ~.robusta subsp. nesophila

PapuaN.G.

120 120

80 80

------40 ~ 40-- ---./

,/~

0 00 40 80 120 0 40 80 120

time (days) --1 ~.macrophylla Aneityum 4 A. robasta subsp. nesophila PapuaN.G..2 ........ " ...... Santa Cruz 5 ~. moorei N.Caledonia -3·· .. ·· .......... Vitu Levu

- 14 -

On Palawan a good tree is said to produce 16 kg of resin per year, butthis could be appreciably increased with more frequent visits and theaverage is only 2 kg. Max~um yield comes from trees of 0.4-0.8 mdiameter. Prices quoted locally are 1.5 peso/kg in Palawan and 2.0 pesoin Manila; but 3.0 peso if smuggled to Borneo (£88, £118 and £176 pertonne respectively). Hence much resin goes to Borneo and official figuresfor Palawan resin production are believed to be 50-100 percent too low(£1 = 17 peso). The quality of resin and its value are said to varythroughout the Philippines (S.C. Halos), and typical buying prices arequoted by her as, Palawan 0.96 peso/kg, Zambales 0.90, Quezon 0.65,Samar 0.30 - the quality is said to differ.

A detailed study has been commenced by S.C. Halos at the new PhilippinesForest Research Institute to investigate difference in resin yield betweentrees and localities and to link this with morphological traits.

Consideration to tapping methods has been given in Indonesia. Sumantriand Sastrod~odjo (1977) discovered that traditional tapping m~thods

were cheaper .and just as high in yield and quality as' others. Malinka(no date) has published simple advice on tapping methods for Irian Jaya.

SEEDLING RESPONSE TO LIGHT

The accumulated info~ation on the ecology of Agathis and its performanceunder different si1vicultura1 regimes clearly shows that seedlings ofsome species and provenances can grow in open, brightly.lit sites andothers need same degree of shading from direct sunlight, and concomitanthigher aerial humidity, lower soil surface temperature and leaf temper­ature. It is easier and cheaper to establish seedlings in the open,andif they can tolerate these conditions they are likely to grow fasterthan in shade. The two extreme responses are exemplified by Malayan(and perhaps all) ~. borneensis which requires shade and grows slowlyand ~. macrophylla, especially the southern New Hebrides provenance,which grows fast in the open.

A start has been made to analyse these responses with the end in view. ofmaking recommendations for silviculture of small Agathis.

An exploratory experiment was conducted in a glasshouse at Wytham nearOxford during the summer and autumn months of 1979 when insolationapproximates tropical intensity although day length is greater and bothtemperature (20-32(-38)oC) and relative humidity «20-)40-85 percen~)fluctuate more than within the humid tropics. Conditions at this timeof year are a very rough approximation to those in the open in the humidtropics.

Seedlings of three provenances of ~. macrophylla (Santa Cruz, Vitu Levuand Aneityum), ~. moorei (New Caledonia) and ~. robusta,subsp. nesophila(Papua New Guinea), were grown in pots in unshaded, lightly ~haded (c. 60percent of unshaded. light) and heavily shaded (c. 35 percent) conditions,from August through to the end of November. Owing to a shortage ofstrictly comparable material no attempt was made at a comprehensiveanalysis of growth, which was assessed simply as stem height and on asubsample as total axis length. All species and provenances grew

- 15 -

fastest without shade o Agathis macrophylla Aneityum provenance seedlingsinitially grew fastest, possibly because they were initially the smallest,but later the growth rate was comparable to the other provenances of~. macrophy11a and ~. robusta subsp. nesophila. ~. moorei consistentlymade less height growth, Fig. la. Similar results were obtained for totalshoot growth.

The growth of all the plants decreased with shade (Fig. 1 b-f), thoughthere are indications of a differential response between species andprovenances. A form factor relating leaf area to length and maximumwidth was calculated (D. Alder). This only ranged from 0.64 to 0.68despite the markedly different leaf shape between species. It isplanned to conduct further glasshouse and growth chamber experimentson seedlings to elucidate some of the basic characteristics of growth andthen to extrapolate to studies in shade h~uses and the forest at a suit­able location in the tropics. Thus we plan to link observations inclosely controlled, relatively uniform environments with others in semi­natural and natural ones. There are too many variables to make it easyto interpret growth in the forest without such back-up analyses forsimpler environments.

At Bogor experiments on A. darnmara seedlings showed that leaf productionincreased considerably with increasing light up to 53 percent full lightand decreased as illumination was increased further. It w~s also stimu­lated by foliar application of nitrogenous fertiliser (Biotrop, 1976a).Work was also begun on competition with Imperata cylindrica, anddifferent shade intensities on transplanted seedlings of the same species(Biotrop) 1976b).

INTERNATIONAL SPECIES/PROVENANCE TRIAL

An international species/provenance trial was established in 1979.During January-March nine countries were visited and seed was eithercollected at the time or arrangements made to have later collectionssent to the U.K. for distribution. In all six collections were received,most from the humid tropics, rather than more strongly seasonal climates.These seeds were sent out in June and July 1979 to seventeen countries(see Table 3) chosen to include a whole variety of site types over alarge geographic range, with the objective of obtaining as much data aspossible on the potential of Agathis as a plantation tree. Table 3 alsoincludes details of direct country to country seed exchanges made during1979.

In many countries through its natural range a heavy and increasingexploitation of rain forests is endangering the wild populations, andequally important consideration in selecting the trial recipients wasgiven to choosing locations for the ~ situ conservation of provenancesand species.

The limited size of the collections (largely resulting from attacks bythe seed-eating larva of Agathiphaga) and continuing problems of seedtransport and storage necessitated unequal distribution among the parti­cipant countries.

- 16 -

As the young trees of many provenances are intolerant of full sunlightit has been 'suggested that the trial be planted with a 'nurse' crop ofLeucaena leucocephala (Philippines, giant variety, or in pne' case exNicaragua) for the first few years of life. Experience in Java and thePhilippines suggests that the association of the two genera is satis­factory and initial growth of the Agathis is promoted. A unit block of49 plants (7 x 7) is suggested at 4 x 4 m spacing, interspersed with asimilar number of pre-established Leucaena, also at 4 x 4 ffi, to give aformal plant. espacement of 2 x 2 m, and a block size of 28 x 28 m. Seedviability was calculated prior to despatch and seed was sent in unit lotsto contain approximately 100 germinable seeds, i.e. sufficient to plantone unit block~ with adequate reserves. - -

After germination it was recommended that the seedlings be pricked outinto 75-120 mm polythene pot~ and gradually hardened off under lightshade over the next 9-18 months. After this time they should haveobtained a height around 0.3 m and be ready for field planting.

The period in the nursery is not critical, and a varying time scale canbe used to fit in with site. preparation and suitable planting weather.Planting into bare ground at a freely draining site, on a soil of goodphysical structure is recommended. Frequent weeding is required in theearly years. Grasses are known to be particularly inhibiting. TheLeucaena nurse should be removed after some five years, by which tUnethe Agathis is likely to have overtopped it.

- 17 -

Agathis in ternationa1 species/provenance trial, 1979

~~•..-4..c

~ "'"0 (1) (1)fJ) rn rnQ) ~ ~d ::s ::s

d s::. ~ ~

0.. cofJ) (1) as co co CG

co co co ..0 c: (1) ...-4 rn .....r-4 r-4 r-I ::s ·M co ..c ~ co Q) eur-I

"'"'"r-I, fJ) :::s ·M fJ) (J ::s .c c: .c co

0 s:: •..-4 "0 co ..... ::s..c ..c ..c:: N <tS ctS 0 r-4 d .u J.c (1) 0.. (1) 00::s .u :3 J.c ..... "0 co ClS fJ) ::s CJ 0.. CJ tU0 0 § a ~ fJ) (1) oJ Q) <U ~~ a a ..... 0 ~~ J.c ,... CJ ::SZ

~s:: ,...~ .u co ,... a (Jr-4 CJ co

0 CJ >-. CJ ..0 eu a C'Q fJ) Q) (J J.c ::s ..... ::s CJco ro .u ctS ctS o co tlS :3 au ::SN -r-4 +J Q).c Q) .....e ...-4 S ..... S +J ,... ::s "0 co e co e co ...-4 P-c ...-4 Z....., Q) d 0.. "'"'" :3 ~

o eM • c:: . ~ • eu o CO • Q) o Q)

<I <I • >< o >=<I~ <1< <100 <IP-c <I Pot <lz <IZ ~I Q) .-11 Q)

Australia 1 1 1 1 6

Brunei 1 1 6

Cameroon 1 6

Colombia 1 6

Costa Rica 1 1 1 1 1 6

Fiji 4

Ghana 1 6

Indonesia 1 6

Ivory Coast 1 1 1 6

New Caledonia 1 4 3 6

New Hebrides 2

New Zealand 2

Nigeria 1 7

Papua New Guinea 2

Philippines 1 1 6

Puerto Rico 1 1 1 6

Reunion 2

West Malaysia 1 1 1 1 1 6

Sabah 1 6

Sarawak 1 6

Sierra Leone 1

Solomon Islands 2

South Africa 1 1 5 5 6

1. Seed despatched via Oxford 1979 4. Seed sent directly from Solomon Islands2. Seed sent directly from New Caledonia 5. Seed sent directly from Australia3. Seed sent directly from New Zealand 6. + 7. Leucaena leucocephala seed to be

used as a nurse crop

- 18 -

REFERENCES

Anon.· 1976. Trials with exotic conifers in the Juan Vinas area ofCosta Rica - prel~inary results. Activities at Turrialba 4,6-7.

Biotrop. 1976a. Effects of light intensity and fertiliser on the growthof Agathis loranthifolia Sa1isb. seedlings by Mr. FlorendoBarangan. Biotrop Newsletter 15, 8.

Biotrop. 1976b. Some effects of shading and competition with Lmperatacylindrica on transplanted seedlings o_f Agathis loranthifoliaSalisb. by Mr. Sulong IbrahLm. Biotrop Newsletter 15, 10.

Bowen, M.R. and Whitmore, T.C. 1980. Agathis, a genus of fast growingrain forest conifers. Commonwealth Forestry Revie~. In press.

Brown, W.H. 1951. Useful plants of the Philippines. Deparbment ofAgriculture and Natural Resources Technical Bulletin 10, Manila.

Brunig, E.F. 1974. Ecological studies in the kerangas forests ofSarawak and Brunei. Kuching.

Dumbleton, L.J. 1952. A new genus of seed-infesting Micropterygidmoths. Pacific Science 6, 17-29.

Hocking, G.H. 1935. A note on the germination of some native species.N.Z. J. For., 3, 225-7.

Hutchins, D.E. 1918. Waipoua Kauri forest, its demarcation and manage­ment. Dept. Lands and Surv.! We11ingto~, N.Z.

Kauri Management Unit Auckland Conservancy. 1978. Kauri volumes (secondcrop). Auckland.

Leakey, R.R.B. and Longman, K.A. 1978. First Annual Report of theMinistry of Overseas Development Research Scheme R3454. Instituteof Terrestrial Ecology. Edinburgh.

Malinka (no date). Damar putih. Penerbit Pustaka Rakjat, Ho11andia.

McKinnon, A.D. 1936.australis) seed.

Collection and germination of Kauri (AgathisN.Z. J. For., 4, 129.

MacMorran, A.M u 1946. ExperUnental seed storage.Hort. 15, 24-25.

J. Roy. N.Z. Inst.

Mirams, R.V. 1957. Aspects of the natural regeneration of the Kauri(Agathis australis Sa1isb.). Trans. Roy. Soc. NuZ., 34, 1661-80.

Mamose, Y. 1978. Vegetative propagation of Malaysian trees. MalaysianForester 41, 219-23.

- 19 -

Nik1es, D.G. 1965. Aspects of the reproduction of Araucariaceae inrelation to phylogeny and tree breeding. Proposition of thePh.D. Committee. University of North Carolina.

Preest, D. 1980. Seed storage of several New Zealand indigenous trees.Part 1 - Kauri (Agathis australis). N.Z. J. For. Sci., 9,337-43.

Sando, C.T. 1936. Notes on Agathis australis. N.Z. J. For., 4, 16-21.

Sumantri, I. and Sastrod~odjo, S. 1977. Percobaan penyadapan Agathis1oranthifo1ia Salisb. di Banyumas T~or. Laporan LembagaPenilitian Rutan 84.

Suriamihardja, S. 1979. Seed characteristics of Agathis loranthifolia.Malaysian Forester 42, 214-20.

Whitmore, T.G. 1977. A first look at Agathis. Tropical Forestry Papers11.

Whitmore, T.e. 1980. A monograph of Agathis. Plant Systematics andEvolution. (In press.)

Whitmore, T.C. and Page, C.N. 1980. Evolutionary Lnplications of thedistribution and ecology of the tropical conifer Agathis. NewPhytole 84, 407-16.

Zieck, J.F.U. 1977. Agathis (Kauri Pine) reconnaissance flight, Telefomin,West Sepik Province 18 May 1977. Res. Notes Minor For Prod.Ull/166-l-0.

ACKN~DGEMENTS

This report represents part of the final results of scheme R2881 of theBritish Overseas Development Administration Whose continued support andinterest is gratefully acknowledged. Foresters and biologists have con­tinued to give assistance, without which this venture would have been~possible, at all the places and on various of the topics mentionedin the text.